We really don’t care what statistical method you used

Update: as pointed out in the comments, the amusing error in this article has been “corrected” (or at least, “edited away”). Thanks for your interest.
Update: I note that this article is now “Highly Accessed” ;)

An integrative analysis of DNA methylation and RNA-Seq data for human heart, kidney and liver
BMC Systems Biology 2011, 5(Suppl 3):S4

(insert statistical method here). No, really.

With thanks to Simon J Greenhill and Dave Winter.

Tags: , ,

13 Responses to “We really don’t care what statistical method you used”

  1. huh, really?
    LOL!!

  2. A few minutes ago this was corrected at the source!

  3. Reblogged this on Scientific B-sides and commented:
    Ha Ha! Hilarious! Not that I wouldn’t do similar things with experimental techniques…

  4. P.S. For those who are thinking about my comment as a “report from an affiliated person” – wrong guess. I am just a curious stranger. It was amusing to see the highlighted “avant-garde” phrase in the orriginal article and then watch it vanished in a few minutes…

  5. I was wondering which was this statistical method, so I entered to see the article. I ended up noticing that the responsible for statistical analysis was KZ, which is the corresponding author :S.
    On the other hand, this looks like a smart strategy to grab some attention on your work and have a “highly accessed” article ;-).

  6. Oleg: Although your 2:46 am comment is very remarkable, it did not leave me in the “Who is this guy?”-limbo that you might have believed it did.

  7. I wonder how long it’ll take before it is also fixed in Medline: http://www.ncbi.nlm.nih.gov/pubmed/22784623 :-)

    • Yes and indeed at PubMed Central. I guess this is an issue when documents are simply handed off to other organisations. It would be nice if there were one workflow which generated the final text at all locations…tell him he’s dreaming.

Trackbacks

Follow

Get every new post delivered to your Inbox.

Join 2,202 other followers

%d bloggers like this: